Systems, methods and articles for reading highly blurred machine-readable symbols

Abstract

Systems and methods for robust recognition of machine-readable symbols from highly blurred or distorted images. An image signal representation of a machine-readable symbol element is transformed into a different space using one or more transform operations, which moves an n-dimensional vector of measured light intensities into another n-dimensional space. The types of transform operations may include blur robust orthonormal bases, such as the Discrete Sine Transform, the Discrete Cosine Transform, the Chebyshev Transform, and the Lagrange Transform. A trained classifier (e.g., an artificial intelligence machine learning algorithm) may be used to classify the transformed signal in the transformed space. The types of trainable classifiers that may be used include random forest classifiers, Mahalanobis classifiers, support vector machines, and classification or regression trees.

Description

BACKGROUND[0001]1. Technical Field[0002]The present disclosure relates to reading or decoding machine-readable symbols.[0003]2. Description of the Related Art[0004]Machine-readable symbols encode information in a form that can be optically read via a machine-readable symbol reader or scanner. Machine-readable symbols take a variety of forms, the most commonly recognized form being the linear or one-dimensional barcode symbol. Other forms include two-dimensional machine-readable symbols such as stacked code symbols, and area or matrix code symbols. These machine-readable symbols are typically made of patterns of high and low reflectance areas. For instance, a one-dimensional barcode symbol may comprise a pattern of black bars on a white background. Also for instance, a two-dimensional symbol may comprise a pattern of black marks (e.g., bars, squares or hexagons) on a white background. Machine-readable symbols are not limited to being black and white, but may comprise two other colors...

AI Technical Summary

Problems solved by technology

Noise and blur can reduce the quality of an image depicting a machine-readable symbol.

An image can be blurred because, for example, the image capture device used to capture the image was out of focus, the image capture device moved while the image was being captured, the object including the machine-readable symbol moved while the image was being captured, the shutter-speed was too slow or too fast for the light conditions, the lens of the image capture device was dirty, etc.

Using conventional techniques, decoding machine-readable symbols from highly blurred or distorted images can be difficult or impossible.